PyTorch
Transformers
computer-vision
masked-autoencoder
age-estimation
facial-analysis
arxiv:2506.14909
Instructions to use fridolin22/FAHR-Face with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use fridolin22/FAHR-Face with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("fridolin22/FAHR-Face", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Interactive Web Demo for FAHR-Face: Age, Frailty (Δ-Age) & XAI Heatmap
#1
by sdCarr - opened
Hi everyone!
I built an interactive Gradio Space demo for the FAHR-Face models published in The Lancet Digital Health (2026).
Features included:
• Biological vs Chronological age gap (Δ-Age) analysis
• XAI Attention Heatmap (visualizing which facial regions drive the prediction)
• Masked Autoencoder (MAE) inpainting / reconstruction demo
• Phenotypic similarity comparator
• Image quality auditor (lighting and sharpness check)
• Downloadable health assessment report
You can try it directly in the browser here (includes public domain demo faces):
👉 https://huggingface.co/spaces/sdCarr/fahr-face-age
Feedback and suggestions are welcome!